To generative: here are 6 principles to use it in journalism

It is known that artificial intelligence suffers from rooted problems when it comes to diversity: automatic learning algorithms are trained on historical data that can incorporate institutional discrimination. The NLP and theTO THEGenerative suffer from the same problems and the industry itself has to face a challenge linked to diversity.

It is therefore surprising that the emerging discussion in the sector on generative artificial intelligence has so far not managed to face these problems: the press release of the British regulator ofcom at the beginning of this month does not mention it; Nor do the new guidelines of the US Radio Television Digital News Association do so. The lessons ofBuzzfeedto and the discussion on its use of technology do not touch the topic; Despite being burned by technology, Cnet does not mention prejudices in his policy.

Wiredmentions prejudices in its declaration on the use of generative artificial intelligence, but only with reference to the generation of content, not in the sections on the generation of ideas or on research. The closest thing I saw coming from an organization is the mention of the principles recently published by the Guardian on "Protection from prejudices" in general terms. Take a look at the main concerns about the artificial generative intelligence identified in a sector survey conducted by Wan-Ififra and you will not see that the prejudices will be mentioned.

Six principles for journalistic use responsible for artificial generative intelligence and diversity and inclusion

The six principles for theJournalistic use responsible for artificial generative intelligenceAnd diversity and inclusion aim to provide a starting point to discuss how we can guarantee that the new practices and work flows involving generative artificial intelligence do not repeat the errors of history, as well as ensuring that journalists have some practical tips on how to start experimenting generative tools.

The principles are:

1. Be aware of intrinsic prejudices: It is easy to forget that generative artificial intelligence is intrinsically partial, like any source. We must make sure that ijournalistsstart from this, which leads to ...
2. Incorporates diversity in your suggestions: a good writing of the suggestions is essential for an effective use of artificial intelligence and, just like the goodjournalistsThey know how to ask good questions, they should also know how to write good suggestions, and this includes to ensure that diversity is actively promoted
3. Recognize the importance of original material and references: This is another family journalistic skill: we should always know where information comes from and ensure that these sources do not represent only the strongest voices. For example, to the question "Who are the twenty most important actors of the 20th century?"Chat GPTHe did not apply one color actor; When asked by Winston Churchill or the American founding Fathers, he did not include important critical facts. We would like to reject these responses from human beings, so it is important to do the same with generative artificial intelligence.
4. Report errors and prejudices: artificial intelligence is still learning: it is important to understand how your use is reflected in this.
5. The text generated by Gai should be seen with journalistic skepticism: Once again, here we are only emphasizing a fundamental journalistic trait (I would say that the ability of generative artificial intelligence to generate text will prove to be one of its less useful applications)
6. Transparent be where appropriate: The range of artificial intelligence applications is so wide that it is difficult to say that you should always be transparent, but it should certainly be a consideration.

Source onlinejournalismblog